Efficient Correction of ELANE mutations in Primary HSPCs of Severe Congenital Neutropenia Patients Using CRISPR/Cas9 and rAVV6 HDR Repair Templates

Patients with the rare pre-leukemia bone marrow failure syndrome severe congenital neutropenia (CN) have reduced numbers of neutrophils in peripheral blood (<500/µl) leading to frequent infections and requiring chronic granulocyte stimulating factor (G-CSF) treatment. The majority of patients harbor heterogenous mutations in ELANE, coding for Neutrophil Elastase.

Up to now, the only curative therapy for CN patients that do not respond to G-CSF or with overt AML remains hematopoietic stem cells transplantation with its associated risks. A clinical need for gene therapy for these patients is imminent. We recently described the CRISPR/Cas9 mediated ELANE knockout as a possible gene therapy approach for CN patients with ELANE mutations (ELANE-CN) (Nasri et al. 2019). As an alternative, we wanted to test if specific target therapy for individual ELANE-CN patients could be an option. Here we describe the correction of ELANE mutations using CRISPR/Cas9 to edit the ELANE gene and recombinant adeno-associated virus 6 (rAAV6) to deliver a template for homology directed repair (HDR).

We selected ELANE mutations p.A57V or p.A57T in exon 2, and p.G214R or p.G214RV in exon 5, both known hot spot mutations observed in G-CSF non-responders or in CN/AML patients (Makaryan et al. 2015). We used SpCas9 V3 and chemically modified sgRNA. For exon 2, we choose the highly efficient sgRNA (Nasri et al. 2019) yielding the benefit, that double-strand breaks (DB) that do not result in HDR correction are producing ELANE knockout. For exon 5, we established a sgRNA that produced average 87% (± 6%) editing in healthy donor cells. Two HDR donor template backbones (DTB) were generated. DTB1 is spanning exons 1-3 and DTB2 exons 4-5 of ELANE. Silent mutations were introduced in the repair templates for both ELANE mutations between the cut site and mutation to enhance HDR.

To test the knock-in efficacy, we electroporated healthy donor CD34⁺cells with CRISPR/Cas9 RNP and transduced them with rAAV6 containing the templates at MOI 10⁵. We achieved 34,5% (± 4,5%) knock-in (KI) and 35,6% (± 2,5%) indels for exon 2, or 39,2% KI (± 12,8%) and 18,85% indels (± 4,25%) for exon 5. Edited cells showed high viability, expanded and differentiated well into neutrophils in vitro.

We further applied this approach to primary HSPCs from 4 CN patients harboring selected ELANE mutations. For p.A57, we achieved 14% (±2,3%) KI and 44,7% (±1,9%) indels. For p.G214, the KI was 59,9% (± 0,1%) and indels 28,8% (± 0,6%). To assess the effect of ELANE correction on the neutropenic phenotype in vitro, we performed CFU and liquid culture neutrophilic differentiation assays. We compared the corrected cells to cells from the same patient that were edited in the AAVS1 safe harbor, as isogenic controls. We observed a significant (p < 0,05) increase in number of CFU-GMs for CRISPR/Cas9 edited HSPCs from two CN patients with p.A57V/T mutations and of CFU-G or CFU-GM for two CN patients with p.G214R/V ELANE mutation. Morphological assessment of Wright-Giemsa stained cytospins of cells derived on day 14 of differentiation revealed significant increases of mature neutrophils for all four edited patient samples ascompared to the respective controls.

Further we performed live cell imaging of neutrophil extracellular trap (NET) formation after PMA stimulation and chemotaxis. NET formation was either improved or comparable between control- and ELANE- edited cells. Chemotaxis showed no difference between control- and ELANE-edited cells. For a patient with p.G214V ELANE mutation, we were able to evaluate chemotaxis and phagocytosis in vivo in zebrafish embryos at 48hpf, as described in Nasri et al 2019. This showed a qualitative improvement of ELANE- corrected cells ascompared to control AAVS1 edited cells. This indicates that our manipulation does not alter the functionality of produced neutrophils while increasing the number of mature cells being produced.

Taken together, we established a protocol for efficient correction of ELANE mutations in primary HSPCs using CRISPR/Cas9 and rAVV6 HDR repair templates. We reached high enough editing to correct the dominant negative effects of mutations, as assessed by markedly improved neutrophilic differentiation in vitro. Generated repair constructs allow fast adaptation to patient-specific mutations in all exons of ELANE. This approach is enticing to be investigated further for clinical translation.